Open-source text-to-speech library whose KittenTTS 2 model clones voices from 5 seconds of audio and runs in real time on a CPU.
16.2K967PythonApache-2.0Updated 4 days ago
thewh1teagle/kokoro-onnx
Python package for offline text-to-speech using the Kokoro model running on ONNX Runtime, with CPU and GPU support.
- Stars
- 2,794
- Forks
- 306
- Open issues
- 80
- Last commit
- 2 months ago
What is it?
kokoro-onnx is a Python package that turns text into speech by running the Kokoro-82M text-to-speech model through ONNX Runtime. It is built on the Kokoro-TTS model and lets developers generate audio locally without calling a cloud speech service.
The project offers several voices and supports multiple languages. Version 1.0 model files are provided for download, and the package is installed from PyPI with pip. A quantized model variant is available, which the project describes as much smaller than the full model, and the author reports near real-time performance on macOS M1.
The repository includes example scripts that save generated speech to a WAV file. It recommends the misaki g2p package for text-to-phoneme conversion with the v1.0 models.
Key features
- 01Runs Kokoro text-to-speech models locally via ONNX Runtime
- 02Installable from PyPI with pip
- 03Offers multiple voices and languages
- 04Provides a quantized model that is much smaller than the full version
- 05Supports both CPU and GPU execution
Use cases
- Adding offline speech output to a Python application
- Generating audio narration for videos or articles from scripts
- Prototyping voice features without a cloud TTS account
Topics
FAQ
- What is kokoro-onnx?
- Python package for offline text-to-speech using the Kokoro model running on ONNX Runtime, with CPU and GPU support. kokoro-onnx is a Python package that turns text into speech by running the Kokoro-82M text-to-speech model through ONNX Runtime. It is built on the Kokoro-TTS model and lets developers generate audio locally without calling a cloud speech service.
- Is kokoro-onnx free to use? What license does it have?
- kokoro-onnx is open source under the MIT license. Check the license terms before using it in commercial products.
- Is kokoro-onnx still maintained?
- Updated within 6 months. The last commit to kokoro-onnx was on Sep 1, 2026.
- What can I use kokoro-onnx for?
- Adding offline speech output to a Python application Generating audio narration for videos or articles from scripts Prototyping voice features without a cloud TTS account
- Where can I get kokoro-onnx?
- The source code and installation instructions are on GitHub: https://github.com/thewh1teagle/kokoro-onnx
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Python inference library for Kokoro-82M, an open-weight text-to-speech model with 82 million parameters.
9.5K1.1KJavaScriptApache-2.0Updated last year
Soprano is an ultra-lightweight, on-device text-to-speech model with CUDA, CPU and MPS support and streaming output.
1.6K138PythonApache-2.0Updated 9 months ago
Desktop and web app that converts EPUB, PDF, text and subtitle files into audio with synced subtitles, using Kokoro-82M TTS.
6.1K500PythonMITUpdated 3 days ago
Open-source voice AI family from Microsoft, with long-form speech recognition (ASR), multi-speaker text-to-speech (TTS), and real-time streaming TTS models.
54.9K6.2KPythonMITUpdated 12 days ago
Open-source text-to-speech system with voice cloning, natural-language emotion control and multi-speaker generation, built on LLM-style models.
33.3K2.9KPythonOtherUpdated 6 days ago